Linguistic Society of America

Stems of the Hittite hi-Conjugation Author(s): Edgar H. Sturtevant Reviewed work(s): Source: Language, Vol. 3, No. 4 (Dec., 1927), pp. 215-225 Published by: Linguistic Society of America Stable URL: http://www.jstor.org/stable/409255 . Accessed: 29/12/2011 05:18
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Linguistic Society of America is collaborating with JSTOR to digitize, preserve and extend access to Language.

http://www.jstor.org

STEMS OF THE HITTITE hi-CONJUGATION

EDGAR H. STURTEVANT
YALE UNIVERSITY

The verbs of the Hittite hi-conjugation fall into three classes according to the final sound of the stem. In spite of more or less irregularity in each class, the following paradigms will represent the usual types of conjugation. I have shown' that certain verbs of the first class (?akki 'he knows', aki 'he dies', ari 'he arrives, has arrived') correspond rather closely to IE perfects with stems ending in a consonant. Most of the other verbs of the same type are etymologically unclear, but from the Hittite point of view they are, with one exception, consonantal stems. We must, however, include in this class the irregular verb au- 'see', whose conjugation is as follows:
Present Preterit Imperative

1 u-hhi 2 au-tti 3 au -zi2 1 au-mmeni 2 au-tteni 3 uw-anzi

u-hhun aug-ta au-mmen

uw-allu au aug-du aug-ten

aw-er

Our second class may be described, from the Hittite point of view, as the a-class. If we assume that these verbs also correspond to IE perfects, we must conclude that a represents an original long vowel, at least in the forms of the singular, since the IE perfect has no stems ending in a short vowel in the first and second persons singular of the active voice. Such orthography as da-a-i 'he takes' and da-a-dg 'he took' is very frequent; but repetition of a vowel sign need not indicate a long vowel. My belief that we have to do with original long vowels in this instance is based solely upon IE grammar.
1 Language 3. 161-8 (1927). 2 This form obviously belongs to the mi-conjugation. 215

216

EDGAR H. STURTEVANT

Hittite a as stem-final of our second class may correspond either with IE &or with IE 5. The latter is to be assumed in the case of da- 'take'. Incompatible as the meanings 'take' and 'give' seem to be, I am conTABLE I
I II III

Present

1 2 3 1 2 3

ar-hi Aagg-ahhi ak-ti akk-i ekk-weni ek-teni akk-anzi ar-hun gagg-ahhun ?ak-ta ek-ta

da-hhi da-tti da-i mema-weni ig?a-teni da-nzi

te-hhi dai-tti da-i tiya-weni tiy-anzi

Preterit

1 2 3

da-hhun

te-hhun hatrae-S pai-g-ta pe-g-ta dai-4 pe-?-ta piya-wen pe-A-ten tiy-er

da-dai-

1 2 3

ekk-wen ek-ten ekk-ir

tarna-ttin da-ir mem-ir da mema-i mem-i da-u da-ttin da-ndu

Imperative 2 3 2 3

6ak pah-i ak-du ageg-ten

da-i pa-u dai-4-ten pe-3-ten tiy-andu

vinced that Hittite da- is the same word as Gk. Lat. d5, etc. tboa6t, has the force as the and Latin Skt. dadami same Greek Although usually has never it 'take' the and it means with when used prefix a, cognates, been clear how that prefix could reverse the meaning of the verb. If,

STEMS OF HITTITE hi-CONJUGATION

217

however, we start with the meaning 'take', the development of the meaning 'give' is easy; 'take' becomes 'choose', and 'choose for so-and-so' is virtually 'give to so-and-so'. A slightly different development is possible from the attenuated meaning 'take (in order to perform some operation)'. The Hittite word, with prefixed sara, often has this force; e.g. Keilschrifttexte aus Boghazkci 4. 12. 1. 20-21 = G6tze, Hattusilis 42: GIM-an-ma-za SES.YA DINGIR.LIM-is DII-at IUr-hi-DUup-an-ma DUMU SES.YA d-ra-a da-ah-hu-un na-an LUGAL-iz-naan-ni ti-it-ta-nu-nu-un, 'When, however, my brother had become a god, then I took Urhi-Tesupas, my nephew, and established him in the kingship.' More important for our purpose is Hattusilis' inaugural address 1. 17f. (p. 8 G6tze): nu-mu A.BU.YA TUR-an ~d-ra-a da-a-d? nu-mu A.NA DINGIR.LIM IR-an-ni pd-es-ta, 'And my father took me, a child, and gave me to the goddess for (her) service.' It would be easy to omit the second verb, and thereby make da- mean 'give'. In one of these two ways, I imagine, IE *d5- got its familiar meaning, while Skt. d-dd- and Hittite da- retain the original meaning 'take'. Very likely some of the verbs in our second class have a from original d, but I do not feel sure of any such etymology. Possibly mema- 'say' is to be connected with Lith. moju, m6ti 'beckon', OS1. na-maja,, -majati 'nod to', Skt. maya 'deception', etc., on the basis of IE *mj(i)-.3 It is at least equally possible, however, that we should think of IE *me'measure'4; note especially Skt. mcti? 'Maas, richtige Erkenntniss', and Gk. TLns'wisdom, skill, counsel'. In the latter case Hittite mema- is a reduplicated perfect with o-grade. In the third column of Table I we quite obviously have various stemforms, and the inconsistency is really much greater than our normalized scheme indicates. The conjugation given in Table I is the most common one-in most respects byfar the most common; but all of the stem-forms are sometimes found outside the categories there assigned to them. Table II, although incomplete, will give some idea of the variety actually found. We must certainly assume a composite origin for the third type of the hi-conjugation. A complete and final solution of the problems involved is impossible at present, but certain facts can be established with the means already at hand. In some respects the most distinctive forms in Table II are the first
3 Cf.

2. 219-20. 1 Cf. ib. 2. 237-8.

Warterbuch derIndogermanischen Vergleichendes Sprachen Walde-Pokorny,

218

EDGAR H. STURTEVANT

persons singular, tehhi and tehhun. This person occurs in the texts less frequently than the third, and consequently it is citable from only a minority of the verbs which are known to belong to this class. We have nehhi 'I turn, send', and nehhun 'I turned'. 'I give' is written pi-ih-hi and 'I gave' pi-ih-hu-un; but the sign BI is used in the value pe as well as pi (e.g. pe-e-di and p6-di for pedis), and the sign AH is employed TABLE II
Stems ending in e e6 ai ai iya

Present

1 2 3 1 2 3

te-hhi ne-(y)atti

peg-ti halzigg-ai

dai-tti (dai)

pais-ti

huwiya-mi halziya-tti tiya-weni

ne-(y)anzi te-hhun unne-A

peg-teni halzigi-anzi

paraiS-teni tiy-anzi huwiya-nnun

Preterit

1 2 3 1 2 3

peg-ta

hatrae-4 paig-ta unnaig-ta dai-4

ighiya-t piya-wen

pe?-ten halze '-ir

nail-tin (nai-r) (dai) tiy-er

Imperative 2 3 2 3 pel-ten pe-(y)andu halzegg-andu

dais-ten tiy-andu

with any preceding vowel. Clearly we should read pehhi and pehhun to harmonize with peiti (pe-es-ti) 'you give', pesta (p6-es-ta) 'he gave', the iterative-durative peklit (p&-e-ki-it) 'he gave', etc. Similarly i?-hi-ih-hu-un 'I levied' (KBo. 5. 8. 2. 3) should be read ilhehhun, although a specific sign for he was available. Few as they are, these forms are the only first persons singular that can be considered original; their
6 See Sommer, Boghazkii Studien 7. 36-45.

STEMS OF HITTITE hi-CONJUGATION

219

only rivals are huwiyami 'I flee', huwiyannun 'I fled', etc., which belong properly to the mi-conjugation. In one case the assumption of an original stem ending in e is supported by a sure etymology. The stem te- 'place' must be identified with IE *dhj-. Orthography gives no hint of a long vowel in tehhi and tehhun, although such forms as da-a-i and da-a-is are common; but, as already suggested, it is hardly possible to interpret the double writing of a vowel as a mark of length, and it is quite certain-that lack of such a doubling does not indicate original short quantity. The contrast between dahhi 'I take': IE *d5- and tehhi 'I place': IE *dhj- proves that original i-stems form one source of our third class. A second source is to be recognized in Pre-Indo-European bases ending in ei. Since these roots yield in the IE languages both a-stems from *mnyo-), the fact that (e.g. eipvrnv)and io/yo-stems (e.g. malivorat the first column of Table II shows e-stems and the fifth iya-stems suggests the presence of the root-final ti.6 Possibly some of the forms in ai(Q) in our third and fourth columns come directly from PIE ei; but we shall see that the forms of the third and fourth columns may be explained by analogy. Several etymologies prove that ji-stems are included in our material. A few months ago I suggested' the connection of huwa-, huwiya- 'flee' with Lat. fugio, etc. At that time I had to assume that the Hittite word came from **bheu-, whereas fugio, Gk. bpeyw, etc. represented a base with root-determinative (*bheu-g-). I am now able to trace both the Hittite and the IE words to an original **bheug"-Brugmann8 assumed a final labio-velar in this root to account for OEng. b7le 'swelling, boil' and OHG b~lla 'blister', and Hirt' connected pebywand on the basis of **bheweg"-. Whether we accept their premises c0ooat or not, they were right about the final sound of the root. For our point it does not matter whether the g of Lat. fugio is due to loss of labialization after u (so Brugmann) or before y (*bhug"y5), as Persson'o would have it. There is no doubt that in Greek the labialization was regularly lost after u". Elsewhere a labio-velar is as satisfactory as a pure velar in this root.
6 The IE languages show fewer io-stems from bases in di and 5i, and bases of these types would not account for the forms in the first and second columns of Table II. SLanguage 3. 114 (1927). 8 Grundriss der VergleichendenGrammatik12. 596. 1 Der IndogermanischeAblaut 135.

10Beitrdge zur Indogermanischen

11 See

Wortforschung 571.

Brugmann-Thumb, Griechische Grammatik4137.

220

EDGAR H. STURTEVANT

Hittite huwa-, huwiya-, then, shows w for original g". That this is the regular development is indicated by wemiya- 'find, get beside Lat. venio, Gk. Pfaic , from *g'myo- 'come'. The two meanings are tied together by Lat. invenio 'find' and Hittite appan wemiya- 'come upon unexpectedly, surprise' (in the military sense). Furthermore wemiyais most naturally translated 'come' in Keilschrift-Urkunden aus Boghazkdi 14. 3. 1. 53. The Hittite king has invited Tawagalawas to come into the city of Mellawanda to meet him, and Tawagalawas has politely refused. The king continues: GIM-an-ma-mu [1. UL .... SAG 'As he did [not] come into the city U]RU an-da zi-e-mi-ya-at .... to me . .. .' Either the Hittite verb contains the full grade of the root or we must assume that syllabic m before i yielded em, although it became am before other vowels and um before consonants"2. Another instance of Hittite w beside IE g" is presented by walh'throw'. That the latter 'strike'3, which is cognate with Gk. aXXco word has an initial labio-velar is shown by Arcadian &7~kXXov-res whether or not we identify it with Skt. galati 'drips, falls'. KOdfXXovrET, For the loss in IE of medial h, which is preserved in Hittite, see my remarks in Language 3. 121 f. (1927). So then, huwiyami is the exact equivalent of Lat. fugio, except for the personal ending, and huwai 'he flees' represents the corresponding a-stem (**bhugW'-),just as dai 'he places' represents an original stem **dhe-. It is no longer necessary to describe the relationship of these two stems as G6tzel' recently did: ". ... huwiyami und huwai, durch ein noch nicht genau fassbares Lautgesetz einander entfremdet. " They were unlike from the first, and their welding into a single conjugational system was a secondary development. A parallel to this word is presented by huwai 'grows', hui-, hui'(i.e. huwi-, huwe?-) 'live'1s. The IE io-stem appears in Lat. Jio 'become' and elsewhere, and the ji-stem is evidenced by Skt. bhavi-tvas'future',
etc.16

In the above mentioned discussion of Hittite ari7 I pointed out the general equivalence in form and meaning between arnuzi 'he brings' and Skt. Troti, Gk. 6pvvvat, between artat 'he took his stand' and Gk. 'he comes' and OP arasam 'I came', and between between arklizzi copro,
See Sturtevant, American Journal of Philology 48. 251-4 (1927). See Sommer and Ehelolf, BoSt. 10. 76 f. and references. 14Zeitschrift fiir Assyriologie Neue Folge 2. 15 (1925). 15 See Sturtevant, Language 3. 110-2 (1927). 16 See Brugmann, Grundriss 22. 3. 153.
12
13

17 Language 3. 161-8 (1927).

STEMS OF HITTITE hi-CONJUGATION

221

ari 'he (has) arrived' and Skt. dritha 'thou art sprung'. I noted also the parallelism in meaning between Hittite arai 'he rises (against)' and Lat. adoritur. We are now prepared to see the connection in form also between these two; it happens that no forms in iya from this particular verb have come to light, but it is safe to say that the first person plural was *ariyaweni or, less probably, a middle form, *ariyawa~tati. The ai-stem of the metaplastic forms warraizzi and warrait indicates that the equivalent warrai should be assigned to the third class of the hi-conjugation. If Hroznfis8 is right in interpreting anda warrai as 'he shuts in', he is also right in comparing Lith. veri. 'open' or 'close' < and Lat. operio 'cover' >. Hittite sai, pl. iyanzi 'put on; seal'"9is to be connected with Skt. syati, sindti 'binds' (perfect sasau). If the Skt. word is related to Lat. saeta 'bristle', OEng. sdda 'rope, snare', etc.20, we must assume an IE root *sai-. Our Hittite verb suggests that there may have been a byeform *sei- (cf. *bhewei- beside *bhewd-). It is possible, however, that a'iwas conjugated according to the second or a-class, as would be natural for a verb with original &i; there would be nothing strange in a miconjugation verb ?iyami from PIE **sdi-; in fact, that is virtually what we have in Skt. syati. A preterit *as' or an imperative, *?a would settle the question in favor of the second class, while a preterit *jai? or *sesta would indicate the third class. Even if we should find a preterit *?ai?, we should have to reckon with the possibility of a transfer from the second to the third class in Hittite itself. Such a transfer must apparently be assumed in the case of peddai 'he flies, flees', peddais 'he fled'21,which betrays the original vocalism of Doric i'rrav 'I flew, fled' in the imperative pe-e-da22. No doubt the transfer to the third class was due to the influence of huwai 'he flees'. Beside nai 'he turns, sends'23we have a third pl. neyanzi and middle
is BoSt. 3, 2163. 19See Sommer and-Ehelolf, BoSt. 10. 36 f.; Ehelolf, Orientalistische Litera-

turzeitung 29. 987 f. (1926). 669. ~O See Walde, Lateinisches Etymologisches W6rterbuch2 21 See G6tze, Hatt. 85; Friedrich, Staatsvertragedes Hatti-Reiches in Hethitischer Sprache 81, 156. 22 KBo. 5. 4. 2. 48 = Friedrich, Staatsvert. 68. Friedrich, op. cit. 156, cites also a preterit pid-da-a-db, but pid-da-a-il stands in the passage to which he refers. 23 See Weidner, Archiv. fiir Keilschriftforschung 1. 62 f. (1923); G6tze, Hatt. 76; Friedrich, ZA NF. 3. 199 (1927). Friedrich, ib. 2. 52 f. (1925), discusses the compounds pennai and unnai 'he drives'; but I doubt his interpretation of the prefixes. More convincing on the prefix pe is G6tze, Hatt. 83 f.

222

EDGAR H. STURTEVANT

forms neya, neyari, etc. While this stem does not harmonize with the iya-stems of other verbs of our third class, it agrees perfectly with Skt. nayati 'he leads'. The Skt. participle nitas requires a dissyllabic and Hittite nehhi, nehhun, nai', etc. indicate that it base **neya&-24, was **ney&-. Occasional forms of the second class (na, pennahhi, pennahhun, etc.) are due to the influence of uta- 'bring', peda- 'bring', and weda'take', all of which are regularly conjugated according to that class. One of the commonest verbs of the third class is pai 'he gives' (pl. piyanzi). If the word has an 1E etymology at all, we must connect it with Skt. pydyate 'be exuberant, swell, overflow'. Other words from the base **poy.-25 are fairly common in the IE languages, and they indicate that the Skt. verb has preserved the original meaning pretty closely. A semantic development: 'be exuberant, overflow' > 'be generous' > 'give' is not improbable. The IE languages do not disclose the quality of the final vowel of the base; but there is nothing to interfere with one's assuming that it was E, as I have done on the Hittite evidence. The proposed etymology involves the assumption that initial py becomes p in Hittite, and until some further evidence to this effect can be adduced, it must remain somewhat doubtful. In view of the approximation of Latin eo 'go' to the form of the verbs of the fourth conjugation, both in the present and in the perfect, it is not surprising to find in our third class a compound verb whose second member is cognate with eo. Friedrich26 has shown that appai (pl. appiyanzi) means 'is finished, is over', e.g. sal-li a-se-es-sar a-ap-pa-a-i, 'der grosse Gottesdienst ist aus'. In this phrase appai is equivalent to Lat. abiit, and so -ai corresponds to iit. The inflection of the perfect of *ei- 'go' in IE presents a difficult problem, and I shall not now attempt to fit the Hittite material into so doubtful a structure; but I feel sure that it belongs there. We may have another compound of *ei- 'go' in Hittite zai 'he crosses'27 (preterit zaiv). If so the prefix is probably the same as in zenna'bring to an end, complete'28,from *ze 'through' (?) and na- 'send'. One naturally thinks of a connection with Gk. &A 'through'. The second and fourth columns of Table II differ from the first and third respectively in the insertion of I immediately before the personal
2i See Walde-Pokorny, VWIS 2. 321. 21See Walde-Pokorny, VWIS 2. 73-75.
28ZA NF. 2. 2942 (1925).
27 See

G6tze, Hatt. 78 and references. 28See Weidner, AKF 1. 64 (1923).

STEMS OF HITTITE hi-CONJUGATION

223

endings. This must be the same element which in the IE languages forms aorist stems. If some verbs prefer the stem in S and others the simple stem, that merely reminds us that in the IE languages some verbs have the sigmatic aorist and others the root aorist, while others still have both. Although the verbs of the third class belong to the hi-conjugation, which corresponds in general to the IE perfect system, it is now clear that they include present and aorist stems as well. It is, however, impossible to trace the three categories in detail. The stem of tehhi 'I place' may be more closely akin to that of the Skt. perfect dadhau or to that of the Skt. aorist adhdt. The preterit nailta, probably pronounced [naist], 'he turned, sent' may be directly comparable with the 'he led', or it may be an analogical creation, as sugSkt. aorist below. anai.It gested We must furthermore beware of assuming that the Hittite verb ever passed through a stage in which present and aorist were as sharply distinguished as they are in some of the IE languages. Quite possibly that distinction was in part of later growth than the separation of Hittite from the parent stock. Even in IE we have presents with aoristic formation. In particular s is a familiar present suffix in IE just as it is in Hittite, and in both it is the prior element of the conglomerate suffix sk29. In general Hittite stems in S which stand beside stems in e may be compared with IE aorists; but halze's- 'call', the commoner stem beside halzai 'he calls', is not essentially different from stems like igtamag- 'hear', mau's- 'fall', and punull- 'ask', which have no vowel stems beside them. In Table II we entered in the third column several forms which do not really have a stem in ai, namely the third sing. pres. dai, the third pl. pret. nair, and the imperative dai. In the first of these i is the invariable third sing. pres. ending of the hi-conjugation. It probably represents e, the IE perfect ending (e.g. Gk. oitE). The consistent Hittite orthography indicates that final e had changed to i, and I do not know of any conflicting evidence. Final e is comparatively rare in the Hittite documents, and seems everywhere to represent an original diphthong; ke 'haec', ape 'illaec', kue 'quae' show final ai of the neuter plural of the pronominal declension, and udne 'countries' seems also to follow that declension. The anomalous a-vd-dg-he (Boghazgdi-Texte in Umschrift 2. 10 -y 23) is merely an erratic way of writing aga hi 'I sit'.
29 Cf. Brugmann, Grundriss 22. 3. 336-52, and, for the Hittite, see Sturtevant, Language 3. 112, 1133' (1927).

224

EDGAR H. STURTEVANT

A comparison of the imperatives da 'take' and dai 'place' suggests that dai is here to be regarded as the bare stem. That, I imagine, is the reason why it has become customary to cite verbs of the third class in this form. When we observe, however, that verbs of the second class show such imperatives as memai 'say' and tarnai 'leave', as well as da and tarna, and that we have from consonantal stems (the first class) not only sak 'know' and the like but also pahs, 'protect', it becomes clear that i is a formative element in the imperative as well as in the indicative. In fact the imperatival i is not confined to the hi-conjugation, as witness kuenni 'strike, kill' beside kuenzi 'he kills'. Imperatival i, then, behaves in about the same way as the IE imperative ending dhi, of which Brugmann30says: 'Der Ausgang dhi diirfte eine partikel sein, die mit dem als Imperativ fungierenden reinen Tempusstamm univerbiert worden war.' With this dhi we must certainly combine the ending t of Hittite imperatives such as it 'go' and wahnut 'turn'. Since there is no indication that Hittite lost final vowels, I suggest that there were originally two independent elements, dh and i, which might be used to strengthen an imperative; Hittite has preserved them both, and IE has amalgamated them to form dhi. But what shall we say of the stem-final of dai 'he places', dai 'place', nair 'they turned', etc.? The obvious answer is that ? became a before i; but of course we cannot derive dai 'he places' directly from PIE **dh~-e. That would have yielded Hittite *te or *ti. A plausible hypothesis is this: after the change of final e to i, the third personal ending of other hi-conjugation verbs was introduced in those of the third class, and then *de-i became dai. The history of the imperative would be parallel, except that the ending was i from the start. In the third pl. pret., however, the ending was er, as is shown by the frequent orthography -e-ir, and it is unlikely that *ne-er would yield nair. Here we must apparently assume analogical influence of the second class upon the stem vowel. This may be the correct explanation of the stem vowel a in the third sing. pres. and in the imperative also. The remaining forms also of the third column of Table II were probably formed on the analogy of the second class, as follows: da : datti = dai : daitti da : dag = dai : dai' The forms of the fourth column may result from a contamination of the second and third columns. Given pe ti 'thou givest' and daitti
30 Grundriss 22. 3. 569.

STEMS OF HITTITE hi-CONJUGATION

225

'thou placest', it is not strange to find also paisti 'thou givest'. It may be, however, that we have here the suffix is' appended to the stem in e, with change of e to a before i. The possibility that we should find here stems in ai from original &i was mentioned above (p. 223). I see no way of deciding between the three explanations. If we are to cite Hittite verbs by their stems, it is difficult to choose between the various stems of the third class of the hi-conjugation. From the historical point of view the stem in e is most important, but for many verbs no forms from this stem are quotable. Perhaps the best plan is to choose the stem or stems that seem most characteristic of eachverb; e.g. te- 'place', pe(?)- 'give', halzes's- 'call'. In doubtful cases it would be better to cite an actual form, e.g. ?ai 'he puts on'. At any rate we should no longer cite a secondary stem like dai- 'place'.